Serological Evidence of Zika Virus Infection in Febrile Patients and Healthy Blood Donors in Sabah, Malaysian Borneo, 2017–2018

ABSTRACT.

Several Zika virus (ZIKV) seroprevalence studies have been conducted in Africa, Asia, Oceania, the Americas, and the Caribbean. However, studies on ZIKV seroprevalence are limited in Malaysia though several studies have shown that the disease is endemic in the Malaysian state of Sabah. To evaluate the seroprevalence of ZIKV infection, 818 serum samples were collected from febrile patients and healthy blood donors from the Kudat and Kota Kinabalu districts in Sabah from 2017 to 2018. They were screened for ZIKV infection by IgM and IgG ELISA, and positive ZIKV IgM samples were subjected to a 90% neutralization test for confirmation. Twenty-four (6% [95% CI 4 to 8]) confirmed and two (0.5% [95% CI 0.13 to 1.8]) probable ZIKV infections were detected among 400 febrile illness patients. Of 418 healthy blood donor samples, six (1.4% [95% CI 0.65 to 3]) were determined as confirmed ZIKV infections and six (1.4% [95% CI 0.65 to 3]) indicated probable ZIKV infection. This is the first study on the seroprevalence of ZIKV infections among patients and healthy blood donors in Sabah. Compared with previous studies in Malaysia, this study shows that the incidence of ZIKV infection has increased. It also suggests that a sero-surveillance system is essential to determine the circulation of ZIKV in Sabah, Malaysia.

INTRODUCTION

Zika virus (ZIKV), a single-stranded RNA virus, is mainly transmitted to humans by the bite of Aedes mosquitoes. First isolated from a febrile sentinel rhesus monkey in 1947 at the research station in the Zika Forest, Uganda,¹ ZIKV has since been circulating at low levels in Asia and Africa. In 2007, a major epidemic occurred on Yap Island, Micronesia, which then spread to French Polynesia, Brazil, and the Americas from 2013 to 2015^2–4 and through Asia in 2016. Zika virus is a flavivirus related to dengue virus (DENV), Japanese encephalitis virus (JEV), yellow fever and West Nile viruses.⁵ It is transmitted to humans through the vector of Aedes mosquitoes or through nonvector transmission such as sexual contact, mother-to-fetus transmission, and blood transfusions.^6–8 Among flaviviruses, ZIKV and DENV share similar symptoms of infection, transmission cycles, and geographic distribution.⁵ Laboratory tests are, therefore, crucial for an accurate diagnosis, especially as it has been reported that an estimated 80% of ZIKV-infected persons are asymptomatic.^9,10

Zika virus seropositivity has been reported in humans, monkeys, and orangutans in Malaysia.¹¹ For example, serological studies in Malaysia and Borneo from 1953 to 1954 detected the presence of ZIKV antibodies in human sera, while ZIKV was isolated from Aedes aegypti mosquitoes in Malaysia in 1966.^12–14 Furthermore, a seroprevalence study from 1996 to 1997 detected ZIKV neutralizing antibodies in 44.1% and 8.5% of sera from humans and wild or semicaptive orangutans, respectively.¹³ In 2014, ZIKV infection was reported in a German tourist who traveled to the Sabah state of Malaysia.¹⁵ An outbreak of autochthonous ZIKV occurred in Sabah in 2016; ZIKV infection was confirmed in two patients who were residents of Kota Kinabalu.¹⁶ However, after the explosive ZIKV epidemic from 2013 to 2016, no seroprevalence study has been conducted on human residents in Sabah. Seroprevalence studies are an effective way of evaluating the true disease burden of flavivirus infections.¹⁷ The aim of the present study was to determine the prevalence of ZIKV infection among patients with febrile illness and healthy blood donors in Sabah from 2016 to 2017.

MATERIALS AND METHODS

Samples.

From 2017 to 2018, serum samples were collected from 400 patients with febrile illness at the Kota Marudu Hospital and 418 healthy blood donors at the Queen Elizabeth II Hospital in Kota Kinabalu City, Sabah. Inclusion criteria for patients were 1) more than 5 years old, 2) mentally and physically healthy, and 3) acute fever. Exclusion criteria for patients were 1) less than 5 years old, 2) mentally and physically unhealthy, 3) suffering from any acute or chronic diseases, and 4) refuse to give informed consent and demographic information. Inclusion criteria for healthy blood donors were 1) more than 18 years old, 2) mentally and physically healthy, 3) no history of hospitalization in the 6 months prior to sample collection, and 4) free from other infections during collection of blood. Exclusion criteria for healthy blood donors were 1) less than 18 years old, 2) mentally and physically unhealthy, 3) suffering from any acute or chronic diseases, and 4) refuse to give informed consent and demographic information.

Ethical approval.

This work was approved by Institutional Ethical Review Committee of the Institute of Tropical Medicine, Nagasaki University, Japan (191003223). The study protocol was registered under the National Medical Research Registry and protocol was approved by the Medical Ethics and Research Committee, Ministry of Health Malaysia (NMRR-19-2898-49362 and NMRR-19-2994-50918).

Viruses and cell lines.

The virus strains used in the ELISA and neutralization assay were MR 766 (ZIKV), 99St12A (DENV-1), 00st22A (DENV-2), SLMC50 (DENV-3), SLMC318 (DENV-4), and JaOrS982 (JEV). To propagate these viruses and apply neutralization tests, C6/36 (Aedes albopictus mosquito cell) and Vero (African green monkey epithelial kidney cell line ATCC CCL81) cell lines were used, respectively.

Anti-ZIKV, DENV, and JEV IgM-Capture ELISAs.

Serological confirmation of ZIKV, DENV, and JEV was obtained by an in-house IgM-capture ELISA as previously described.^18,19 In brief, 96-well microplates were coated with anti-human IgM goat IgG antibody (Cappel ICN Pharmaceuticals, Aurora, OH) for 4°C overnight and were blocked with Block Ace (UK-1 B 80, Yukijirushi, Sapporo, Japan). After washing the plates with PBS Tween (PBS-T), a 1:100 dilution of tested samples and positive and negative controls were distributed into duplicate wells. Plates were incubated at 37°C for 1 hour and then washed as described above. Zika virus or DENV or JEV antigen was added and incubated at 37°C for 1 hour. After subsequent washing, wells were treated with horseradish peroxidase (HRP)-conjugated anti-flavivirus mouse monoclonal antibody (12D11/7E8) at a 1:1,000 dilution for ZIKV Ag or at a 1:1,500 dilution for DENV or JEV Ag. Plates were incubated at 37°C for 1 hour and washed as above. Color was developed by adding o-phenylenediamine dihydrochloride (OPD) (Sigma Chemical, St. Louis, MO) with hydrogen peroxide and an incubated room temperature (RT) for 30–60 minutes in a dark room. A stop solution (1N sulphuric acid) was added, and the optical density (OD) was then read at 492 nm (Multiscan JX, model no. 353; Thermolab System, Tokyo, Japan). A positive control (or test sample) or negative control OD ratio greater than, or equal to, 2.0 was considered positive.

Anti-ZIKV IgG Indirect ELISA.

To detect the presence of anti-ZIKV IgG in serum samples, an in-house anti-ZIKV IgG indirect ELISA was established for this study. A purified ZIKV was applied as an assay antigen in this ELISA system instead of JEV, which was used in the flavivirus IgG indirect ELISA.²⁰ The 96-well microplates were coated with ZIKV antigen (250 ng/100μL/well) and incubated at 4°C overnight. All wells were blocked with Block Ace and incubated at RT for 1 hour. After washing the plate with PBS-T, a 1:1,000 dilution of tested samples, positive and negative controls was added in duplicate wells. After incubation at 37°C for 1 hour, plates were washed and 1:30,000 diluted HRP-conjugated goat anti-human IgG (American Qualex, San Clemente, CA) was added into the wells. These were then incubated at 37°C for 1 hour before being washed. With the addition of an OPD substrate solution (described above), plates were incubated at RT for 30–60 minutes in the dark before the reaction was stopped by adding the stop solution. A standard curve was prepared by using the OD492 values of the positive control serum starting with a 1,000-fold dilution, followed by serial 2-fold dilutions up to 1:212. Immunoglobulin G titers of patient serum samples were determined from the positive standard curve. A sample titer equal to, or greater than, 1:3,000 was considered positive.

Neutralization test.

To confirm the presence of a specific antibody for ZIKV, DENV, and JEV in febrile patients and healthy blood donors, serum samples were checked for the ability to neutralize ZIKV, the four serotypes of DENV, and JEV using a 90% focus reduction neutralization test (FRNT90), as described previously.^18,19 Serially diluted heat-treated serum samples were mixed with equal volumes of each virus at 60 focus-forming units and were incubated at 37°C for 1 hour for the virus antibody neutralization reaction to occur. Vero cells were infected with a virus and serum mixture. After incubation at 37°C for 1 hour, the infected cells were overlaid with 1.25% methylcellulose in minimum essential medium. The plates with ZIKV or JEV were then incubated at 37°C for 2 days, and the plates with DENV were incubated for 3 days at the same temperature. Once washed with PBS (-), the plates were fixed with 4% paraformaldehyde phosphate-buffered solution and were each permeabilized with 1% NP-40 solution at RT for 30 minutes. Pooled human serum samples with a high titer (≥ 52,000) of anti-flavivirus IgG which recognized DENV, JEV, ZIKV^19,21,22 (diluted 1:1,500) were then added to each well and incubated at 37°C for 1 hour. After washing, 1:1,000 diluted HRP-conjugated goat anti-human IgG was added to each well at 37°C for 1 hour. Subsequently, the staining of positive cells was visualized by the addition of a 0.5 mg/mL solution of substrate 3,3′-diaminobenzidine tetrahydrochloride at RT for 10 minutes, and the number of foci per well was counted using a microscope. The reciprocal of the endpoint serum dilution that provided a 90% or greater reduction in the mean number of foci, relative to the control wells, and that contained no serum was considered to be the FRNT90 titer.

ZIKV case definition.

Following the WHO standard,²³ confirmed ZIKV infection was determined when the serum sample of ZIKV RNA was detected by reverse transcription polymerase chain reaction (RT-PCR) or ZIKV IgM was detected by ELISA. Confirmed positives for the neutralization test were only determined against ZIKV (but not with other flaviviruses) or when the neutralizing antibody titer against ZIKV was ≥ 4 times higher than the antibody titers against other flaviviruses. A probable ZIKV infection was determined if the IgM antibody against ZIKV was detected by ELISA, the ratio of the ZIKV neutralization titer to other flavivirus neutralization titers was < 4 times and if no ZIKV RNA was detected by RT-PCR.

Statistical analysis.

Statistical analysis was performed using R program version 3.4.4 (R Foundation for Statistical Computing, Vienna, Austria). The prevalence rate was expressed according to its 95% CI.

RESULTS

Of a total of 400 serum samples from febrile illness patients, 38 (9.5%) were positive ZIKV IgM, which included 18 (4.5%) for IgM against ZIKV only, 16 (4%) for IgM against ZIKV and DENV, 2 (0.5%) for IgM against ZIKV and JEV, and 2 (0.5%) for IgM against ZIKV, DENV, and JEV. Among 38 ZIKV IgM patients, 36 (94.7%) were positive for ZIKV IgG (Table 1). Out of 418 healthy blood donors, 16 (3.8%) were positive for ZIKV IgM, of which 8 (1.9%), 5 (1.2%), 1 (0.2%), and 2 (0.5%) were IgM positive for ZIKV only, ZIKV and DENV, ZIKV and JEV, and ZIKV, DENV and JEV, respectively. Of 16 anti-ZIKV IgM-positive healthy blood donors, anti-ZIKV IgG was detected in 15 (93.7%) persons.

Table 1.

Antibody profiles of febrile illness patients and healthy blood donors in Sabah, Malaysia, in 2017–2018

Sample collection year	Tested samples	IgM positive ZIKV (%)	ZIKV only (%)	ZIKV & DENV (%)	ZIKV & JEV (%)	ZIKV & DENV& JEV (%)	ZIKV IgG positive (%)
Febrile illness patients
2017	400	38 (9.5)	18 (4.5)	16 (4.0)	2 (0.5)	2 (0.5)	36 (94.7)
Healthy blood donors
2018	418	16 (3.8)	8 (1.9)	5 (1.2)	1 (0.2)	2 (0.5)	15 (93.7)

DENV = dengue virus; JEV = Japanese encephalitis virus; ZIKV = Zika virus.

To confirm the presence of ZIKV infection, the ZIKV IgM positive serum samples were checked for their ability to neutralize ZIKV, four serotypes of DENV, and JEV. Of the 38 ZIKV IgM positive samples from patients with febrile illness, the neutralization titer against ZIKV was at least four times higher than that against all the DENV serotypes and JEV in 24 (6% [95% CI 4 to 8]) samples, which were considered as confirmed ZIKV infections (Table 2). Two other samples (0.5% [95% CI 0.13 to 1.8]) were considered as probable ZIKV infection because the neutralization titer against ZIKV was higher, but not four times higher, than that against all the DENV serotypes and JEV.

Table 2.

Confirmed and probable ZIKV infection among patients in the Kudat district in Kota Kinabalu, Sabah, in 2017–2018

	IgM				FRNT90
Sample ID	ZIKV	DENV	JEV	ZIKV IgG titer	ZIKV	JEV	DENV-1	DENV-2	DENV-3	DENV-4	Diagnostic interpretation*
P-6	35.7	9.5	2.5	51,855	1,280	80	320	160	80	40	ZIKV
P-26	12.3	3.0	0.4	20,907	1,280	< 20	< 20	160	160	320	ZIKV
P-33	2.5	0.9	0.3	13,907	320	40	< 20	40	80	< 20	ZIKV
P-45	24.5	7.3	1.3	69,577	2,560	160	320	320	160	80	ZIKV
P-46	22.0	4.6	0.5	27,150	2,560	< 20	640	320	160	80	ZIKV
P-50	58.1	0.2	0.8	7,632	1,280	44	< 20	< 20	< 20	< 20	ZIKV
P-104	29.4	13.8	0.5	45,758	2,560	320	640	640	320	40	ZIKV
P-106	3.3	2.0	0.5	48,589	1,280	< 20	< 20	160	80	80	ZIKV
P-118	9.9	1.2	0.8	56,103	2,560	40	< 20	320	80	80	ZIKV
P-119	17.0	1.7	4.1	62,311	2,560	80	< 20	80	80	40	ZIKV
P-120	4.3	1.7	0.8	28,058	320	40	< 20	80	< 20	80	ZIKV
P-127	2.4	1.4	0.4	35,568	1,280	40	40	80	40	40	ZIKV
P-192	29.0	1.1	0.6	26,278	1,280	< 20	320	80	40	80	ZIKV
P-207	39.2	0.7	0.8	32,582	1,280	< 20	< 20	< 20	< 20	< 20	ZIKV
P-212	6.3	1.7	0.7	10,816	160	< 20	< 20	40	40	< 20	ZIKV
P-253	5.3	0.8	0.3	4,882	160	< 20	< 20	< 20	< 20	< 20	ZIKV
P-258	17.0	2.5	0.3	47,425	2,560	< 20	640	160	80	80	ZIKV
P-262	15.0	4.0	0.2	39,926	640	160	640	320	80	40	probable
P-267	53.6	7.7	0.9	42,717	640	160	< 20	160	80	< 20	ZIKV
P-288	9.1	3.6	0.1	21,995	320	< 20	< 20	80	< 20	< 20	ZIKV
P-310	28.4	2.4	0.6	29,556	320	< 20	80	40	< 20	< 20	ZIKV
P-319	48.9	12.0	4.3	60,178	1,280	40	320	640	160	80	probable
P-333	15.8	2.7	0.3	42,188	320	< 20	40	80	< 20	40	ZIKV
P-359	10.3	4.4	0.1	22,361	640	< 20	40	160	80	80	ZIKV
P-376	20.7	7.1	0.3	50,517	5,120	< 20	160	320	320	160	ZIKV
P-398	12.1	1.9	0.3	6,880	640	< 20	160	< 20	< 20	< 20	ZIKV

DENV-1-4 = dengue virus serotype 1–4; Diagnosis interpretation* = ZIKV (confirmed ZIKV infection), Probable (probable ZIKV infection). Bold font indicates highest IgM and neutralization value in Table.

Of the 16 positive ZIKV IgM samples from healthy blood donors, 6 (1.4% [95% CI 0.65 to 3]) were confirmed as ZIKV infections as the anti-ZIKV neutralization titers were four times higher than those against DENV-1 to 4 and JEV (Table 3). In six (1.4% [95% CI 0.65 to 3]) other samples from healthy blood donors, the ZIKV neutralization titer was higher, but not four times higher, than those against DENV-1 to 4 and JEV, and was consequently determined as a probable ZIKV infection. The age distribution of the 24 patients with febrile illness (out of 400) and six blood donors (out of 418) with confirmed ZIKV infection was as follows: two (0.2%) were between 1 and 10 years old; four (0.5%) between 11 and 20 years old; three (0.4%) between 21 and 30 years old; seven (0.9%) between 31 and 40 years old; seven (0.9%) between 41 and 50 years old; five (0.6%) between 51 and 60 years old; and two (0.2%) were above 61 years (Figure 1).

Table 3.

Confirmed and probable ZIKV infection among healthy blood donors in Kota Kinabalu, Sabah, in 2017–2018

	IgM				Focus Reduction Neutralization Test (FRNT90)
Sample ID	ZIKV	DENV	JEV	ZIKV IgG titer	ZIKV	JEV	DENV-1	DENV-2	DENV-3	DENV-4	Diagnostic interpretation*
H-132	47.8	30.6	0.4	64,734	2,560	< 20	640	640	640	160	ZIKV
H-180	8.4	2.5	0.4	32,940	2,560	< 20	< 20	40	80	160	ZIKV
H-207	57.4	7.9	0.4	63,740	2,560	< 20	1280	640	320	160	Probable
H-235	2.3	0.5	0.3	5,929	640	160	320	320	< 20	< 20	Probable
H-241	7.9	2.0	0.8	14,092	640	40	< 20	320	< 20	< 20	Probable
H-248	51.6	21.4	9.2	38,050	5,120	2560	320	320	80	40	Probable
H-257	4.4	2.4	0.5	60,874	5,120	< 20	1280	1280	320	< 20	ZIKV
H-295	2.6	1.1	0.7	2,281	40	< 20	< 20	< 20	< 20	< 20	Probable
H-313	3.5	1.0	0.4	9,924	320	< 20	< 20	< 20	< 20	< 20	ZIKV
H-366	8.0	6.6	0.3	17,270	640	< 20	< 20	< 20	< 20	< 20	ZIKV
H-371	10.5	0.3	0.1	18,342	1,280	< 20	40	320	< 20	< 20	ZIKV
H-373	2.3	0.3	0.2	17,025	40	< 20	< 20	< 20	< 20	< 20	Probable

DENV-1-4 = dengue virus serotype 1–4; Diagnosis interpretation* = ZIKV (confirmed ZIKV infection), Probable (probable ZIKV infection). Bold font indicates highest IgM and neutralization value in Table.

Figure 1.

Age distribution in confirmed and probable Zika virus (ZIKV) infections of patients and healthy blood donors in Sabah, Malaysia.

DISCUSSION

Here, ZIKV infection is reported among patients with febrile illness and healthy blood donors after the 2016 ZIKV outbreak in the Malaysian Borneo state of Sabah. In 2016, the first outbreak of autochthonous ZIKV occurred in Sabah where two ZIKV infections were detected. The strains belonged to the Asian lineage.¹⁶ During the outbreak, mosquito samples and a limited number of samples from volunteers and nonhuman primates (Macaca fascicularis) were tested for ZIKV infection by RT-PCR, and all were found negative. Therefore, testing large numbers of samples becomes essential to determine the extent of ZIKV infection among inhabitants.

In this study, 400 serum samples from febrile illness patients from Kota Marudu Hospital and 418 healthy blood donors from Kota Kinabalu were collected. The study confirmed that ZIKV infection was prevalent among both the patients with febrile illness and healthy blood donors. Its findings are in agreement with a previous study conducted from 2012 to 2017 in Kuala Lumpur, Malaysia, where possible and probable ZIKV seropositive rates were 3.3% and 0.6% among the patients and healthy blood donors, respectively.²⁴ As of September 2018, with only eight cases of ZIKV diagnosed in 2016, no further cases of ZIKV infection were determined by the Ministry of Health in Malaysia after testing 2,360 DENV negative serum samples.²⁵ Contrary to the Malaysian results, a number of studies in Southeast Asia in Myanmar (15.8%), Laos (4.5%), Indonesia (9.9%), Cambodia (63%), and Thailand (70.4%) have reported relatively high rates of ZIKV seroprevalence among asymptomatic adults and children.^19,26–29

In the present study of confirmed ZIKV cases, the IgM ratio was higher against ZIKV than DENV and JEV, with some samples indicating the presence of IgMs against ZIKV, DENV, and JEV, all of which belong to the same Flaviviridae family. This could be due to coinfection or sequential infection of the patients by DENV and ZIKV since they are transmitted by the same mosquito species circulating in the same area.³⁰ In addition, it can be challenging to detect serum showing only ZIKV seropositivity as more than 90% of adults in Malaysia are seropositive for DENV.³¹ Consequently, an in-house ZIKV/DENV/JEV IgM-capture ELISA was used, followed by 90% neutralization tests against ZIKV, four serotypes of DENV, and JEV.^18,19 The neutralization test undoubtedly plays an important role in confirming ZIKV infections in this situation by determining ZIKV IgM cross-reactive with other flaviviruses to rule out ZIKV infection. Zika virus IgM persists for up to approximately 12 weeks,³² and positive ZIKV IgM was detected in apparently healthy people. Moreover, the infection can be mild, meaning that infected people may have failed to develop symptoms and the population developed immunity. A recent study suggested that seroprevalence existed among workers migrating from Southeast Asia to Taiwan who had positive IgM and neutralizing antibodies against ZIKV, and 1% were confirmed as infected with ZIKV.¹⁷ In short, the incidence of ZIKV infection may have been underestimated in Southeast Asia.

Among the mosquito-borne flaviviruses, JEV and ZIKV are endemic in Malaysia, and DENV is hyperendemic with high annual incidence and seroprevalence.^33,34 In this study, anti-ZIKV IgG was detected in 94.7% of patients and 93.7% of healthy persons among ZIKV IgM-positive samples. Higher age groups (31–50 years old) and males were generally the most affected, and the same demographics were also associated with DENV infection.³³ These results may indicate increased exposure to Aedes mosquitos because of the increased time of older age and typical gender behaviors among males.²⁴ Compared with other Asian countries such as Singapore,⁴ Thailand,²⁹ the Philippines,¹⁷ Cambodia,²⁸ Indonesia,²⁷ and Myanmar,¹⁹ the seroprevalence rate of ZIKV infection was low in Sabah, Malaysia. Possible mechanisms could be differences in viral strains, host factors and limitations in the surveillance system.³ Furthermore, another study suggested that the American strain of ZIKV seemed to transmit efficiently through Aedes aegypti; in Kota Kinabalu, however, the Aedes albopictus is the predominant Aedes mosquito.³⁵ ZIKV was not detected in 255 Aedes albopictus collected around ZIKV-positive patients’ residences or in 37 contacts of patients in Sabah in 2016.¹⁶ Our study has limitation that we could not perform all participants to check ZIKV IgG ELISA. We had screened all samples for ZIKV IgM ELISA and anti-flavi IgG ELISA in Sabah, and only ZIKV IgM positive samples are available in Japan (due to limited serum amount and limited number of specimens transportation) to confirm ZIKV IgM ELISA and neutralization tests.

The scope of this study encompassed testing samples from two locations in Sabah, incorporating sources of samples from both febrile patients and healthy blood donors. Additionally, a high neutralization titer was detected among patients and blood donors though a high cut-off value of 90% FRNT was used. In conclusion, the study evaluated the seroprevalence of confirmed ZIKV infection among febrile patients (6%) and apparently healthy persons (1.4%) in 2017–2018 in Sabah, Malaysia. In addition, our study size represented 0.51% (418/77,376) and 0.07% (400/572,500) of total population in Kota Marudu and Kota Kinabalu, respectively. To the best of the authors’ knowledge, this is the first seroprevalence study of ZIKV infection among patients and healthy persons in Sabah. The study suggests that the incidence of ZIKV infection has increased relative to levels reported by previous studies^24,25 in Malaysia after the 2016 outbreak. Asymptomatic infection remains a key epidemiological parameter that can significantly influence disease dynamics, especially estimation of overall attack rates and the level of herd immunity. Asymptomatic infections in healthy persons are acting as a silent driver of the pandemic. There is a low risk of transmission from asymptomatic persons, they might still present a significant public health risk. Our data contributed the scientific community that basic scientists and clinicians need to consider ZIKV as a differential diagnosis in febrile illness patients. Moreover, surveillance studies, as well as knowledge of the ecological and epidemiological conditions, are needed to further determine the circulation of ZIKV in Sabah, Malaysia.